Impact of end-of-life respiratory modalities on quality of dying and death and symptom relief in patients with interstitial lung disease: a multicenter descriptive cross-sectional study

Table 2 Results of multivariate regression analyses of scores of the Good death inventory for quality of dying and death for ILD patients treated with HFNC, NIV, or IMV, compared with those with COT (reference group)

	Average score of 18 domains of GDI			"Physical and psychological comfort" domain score
	B	95% CI	p value	B	95% CI	p value
Respiratory modality (vs. COT)
HFNC	0.49	0.09 to 0.88	0.02	1.43	0.59 to 2.26	0.001
NIV	0.34	−0.22 to 0.91	0.23	0.76	−0.43 to 1.95	0.21
IMV	0.01	−0.75 to 0.77	0.98	0.05	−1.56 to 1.66	0.95
Opioid use (vs. no use)	0.38	−0.05 to 0.80	0.08	-0.13	−1.03 to 0.77	0.77
Sustained sedation (vs. no use)	0.10	−0.36 to 0.55	0.67	0.28	−0.66 to 1.23	0.55

The scores of GDI for QODD for patients treated with HFNC, NIV, or IMV were tested using a multivariate linear regression model with those treated with COT as the reference group. This model included the patient’s age at death, patient’s sex, the relationship between the patient and the family, cause of death (i.e., acute exacerbation or others), respiratory modality, opioid use, and sustained sedation use as independent variable
B, partial regression coefficient; GDI, Good death inventory; HFNC, high-flow nasal cannula; COT, conventional oxygen therapy; NIV, non-invasive ventilation; IMV, invasive mechanical ventilation

ISSN: 1465-993X